Fluency Can Bias Masked Priming of Binary Judgments: Evidence From an All-Nonword Task

*Research article by Glen E. Bodner and Jeremy C. S. Johnson, University of Calgary; Michael E. J. Masson, University of Victoria

Abstract

Using a novel all-nonword task (“Does the target have more vowels than consonants?”), new evidence is provided showing that processing fluency can bias masked priming of binary judgments. Two experiments revealed masked repetition priming for “yes” nonwords (e.g., NUISO) but not for “no” nonwords (e.g., RULON). This pattern is considered evidence that the greater ease of target processing induced by repetition primes was attributed to the target being a member of the response category associated with shorter RTs, namely the “yes” response category. This pattern of nonword priming effects reinforces Bodner and Masson’s (1997) key claims: (a) a fluency-attribution process can influence masked priming, and (b) therefore, the masked-priming paradigm does not isolate lexical processes.

Keywords: fluency bias, lexical decision, masked priming, nonword priming

Executive Summary

Responses to targets can be facilitated by brief masked primes of which participants are unaware. Consistent with a lexical basis for priming (e.g., Forster & Davis, 1984), masked repetition (vs. unrelated) primes typically facilitate lexical decisions for words but not for nonwords (e.g., Forster, 1998). Building on earlier work, Bodner and Masson (1997) argued that masked priming for nonwords might be obscured in the lexical-decision task by a fluency bias, whereby an increase in processing speed on repetition trials might be attributed to the target being a word, creating a bias that works against making correct responses to nonwords (e.g., Balota & Chumbley, 1984; Feustel, Shiffrin, & Salasoo, 1983).

Processing fluency has been shown to affect a variety of binary judgments (e.g., Jacoby, 1983; Jacoby & Dallas, 1981; Whittlesea, 1993; Whittlesea, Jacoby, & Girard, 1990). When the processing of a stimulus is experienced as fluent, observers seek to attribute that fluency to a source. In a recognition task, fluency is often attributed to prior study. For example, masked repetition priming of targets elevates recognition claims (Jacoby & Whitehouse, 1989; Rajaram, 1993). Fluency can be attributed to other task salient sources too, ranging from visual clarity (Jacoby, Allan, Collins, & Larwill, 1988) to fame (Jacoby, Woloshyn, & Kelley, 1989).

*To read the full article please search for Dr JJ's publication online